A comparison of techniques for handling missing data in scenarios with different missing data mechanisms

dc.contributor.advisorBeretvas, Susan Natashaen
dc.contributor.advisorHolahan, Carole Ken
dc.creatorLi, Xiaoyinen
dc.date.accessioned2016-01-29T15:32:40Zen
dc.date.accessioned2018-01-22T22:29:22Z
dc.date.available2016-01-29T15:32:40Zen
dc.date.available2018-01-22T22:29:22Z
dc.date.issued2015-12en
dc.date.submittedDecember 2015en
dc.date.updated2016-01-29T15:32:40Zen
dc.description.abstractThe purpose of this study was to illustrate the influence of missing data mechanisms on results of a multiple regression analysis and to demonstrate the influence of the use of traditional techniques (including listwise deletion, pairwise deletion and mean substitution) versus use of multiple imputation (MI) for handling missing data. A methodological approach involving a real generated dataset was adopted. Results from descriptive analyses and multiple regression models indicated that traditional missing data handling methods and MI yield similar regression coefficients and standard error estimates. Although the means and correlations are almost always biased regardless of missing mechanism and missing data techniques, the bias was less severe when using MI.en
dc.description.departmentStatisticsen
dc.format.mimetypeapplication/pdfen
dc.identifierdoi:10.15781/T2R05Sen
dc.identifier.urihttp://hdl.handle.net/2152/32863en
dc.language.isoenen
dc.subjectMissing dataen
dc.subjectMissing mechanismen
dc.subjectMultiple imputationen
dc.titleA comparison of techniques for handling missing data in scenarios with different missing data mechanismsen
dc.typeThesisen

Files